We estimate arbitrage-free term structure models of US Treasury yields and spreads on BBB and Brated corporate bonds in a doubly-stochastic intensity-based framework. A novel feature of our analysis is the inclusion of macroeconomic variables - indicators of real activity, inflation and financial conditions - as well as latent factors, as drivers of term structure dynamics. Our results point to three key roles played by macro factors in the term structure of spreads: they have a significant impact on the level, and particularly the slope, of the curves; they are largely responsible for variation in the prices of systematic risk; and speculative grade spreads exhibit greater sensitivity to macro shocks than high grade spreads. In addition to estimating risk-neutral default intensities, we provide estimates of physical default intensities using data on Moody's KMV EDFs™ as a forward-looking proxy for default risk. We find that the real and financial activity indicators, along with filtered estimates of the latent factors from our term structure model, explain a large portion of the variation in EDFs™ across time. Furthermore, measures of the price of default event risk implied by estimates of physical and risk-neutral intensities indicate that compensation for default event risk is countercyclical, varies widely across the cycle, and is higher on average and more variable for higher-rated bonds.